A few weeks ago, it was announced that a British government Department was going to be forbidden by the Treasury from undertaking any capital expenditure without prior permission. Apparently this ban had been deemed necessary because of past failures on the part of this Department to prove that its spending represented value for money.
Controversial decision! Not least because this planned capital spending included the grants to renovate social housing which was in such bad repair that mould was causing fatal respiratory illnesses in children. And also a decision which laid bare some complicated facts about control and power.
Most obviously, the constant power struggle between the rest of the civil service and the finance ministry. Also, the hidden power of the “value for money” exercise, in which important decisions are concealed under innocuous looking assumptions. But perhaps less obviously, there is much significance to the fact that only *capital* spending was going to be made subject to this constraint. What is special about capital expenditure versus current, and how are they defined?
The answer’s not important; I can talk about it if you care but the important thing is that it’s a stipulation. At some point in the past, a convention has been adopted, by sensible people who understood what they were doing, to the effect that presenting numbers in *this* way rather than *that* way would give the best chance of making the right decisions.
That’s what an accounting system is; a set of principles for organising information for the purpose of decision making. And a key problem – perhaps the key problem – for this particular kind of organising system is the lack of a natural unit of time.
Back in the days when double entry was invented, accounting for a merchant venture was a matter of recording the cost of what you bought, the price you sold it for, and your expenses along the way. The time period for the accounts was the length of the journey there and back; there was no interim reporting while the caravan was out on the road or the ship at sea.
And there were no interim decisions to make either: once your capital was invested in a mercantile venture, it was out of your hands, and you had no more decisions to make until the caravan returned. (At the beginning of “Merchant of Venice”, Antonio doesn’t know he’s broke). The accounting period was *matched* to the time horizon of the decision making process.
For anything more complex, which persists over multiple periods and where you can change your mind at any moment, using the double-entry concept requires you to distort things, like squashing a globe onto a flat map. If you are doing more than one thing, you have multiple control horizons; a single system for reporting will necessarily lose information.
Hardly anything is truly “current” expenditure. Almost everything any organisation does is meant to have at least *some* lasting effect.
A lot of the time, the stipulations are made for reasons of convenience for the accountants rather than accuracy. Because managing information is expensive, and in this context “convenience” means “possibility”.
And the reduction in information is not even necessarily a problem. The information has to be reduced, in order to, literally, make it manageable. Selecting what you will pay attention to is, necessarily, selecting what you are going to ignore. (Stafford Beer said that “ignorance is the information processing system of last resort”).
But things change, and the circumstances under which information bargains and compromises were agreed in the past may not be those of the present.
This is a problem, because the accounting system is the system which represents information to the decision maker. It cannot be assumed that it will communicate the fact that it has ceased to accurately represent. This piece of information might not be something that it is capable of expressing.
In too many cases, you find out that the accounting system has drifted hopelessly from the reality it was meant to describe in “the bad way”. Something physical (ie, something outside the accounting system) happens, which is bad, impossible to ignore, and totally unexpected.
It might be better to review assumptions ahead of time, and spend effort on understanding what kind of information is being attenuated.
Plenty of food for though here. I’m reminded of our mutual old friend the bond “market” and rules about marking values to it. This week, what with Credit Suisse, and the largest pension fund in Sweden losing 8% of its equity portfolio in SVB et al, it seems some bankers and supervisors were blindsided by out of the ordinary effects when rates moved from 15 years at nil to normal in a short period. The real world happens in the transitions between one set of parameters (or accounting periods) and another, as Minsky understood but accountants often don’t seem to.
Dead Souls by Gogol is the ultimate arbitrage of accounting periods.